METHODS USED IN THE STUDY OF STARCHES. 297 



recorded the form and size of the grain; the position and form of the hilum, or the assumed 

 Y)omt of origin of growth or center of organic structure; the form, number, and other char- 

 acteristics of the kimella" ; the characteristics pertaining to the form of the grains, whether 

 singly or in doublets, triplets, or aggregates, etc. Many of the miimter features of the 

 grains that were observed will not for obvious reasons be seen in the photograi)hic repro- 

 ductions. In describing the grains the terms "proximal end" and "distal end" have 

 been adopted, the former being the end nearer which the hilum is located. The "longi- 

 tudinal axis" corresponds with an imaginary line, extending from the proximal end through 

 the hilum to the distal end. In different starches and in different grains of the same kind 

 of starch this may be the long or the short axis. The measurements of eccentricity of 

 the hilum have reference to the distance of the hilum from the proximal end of the 

 longitudinal axis. 



IODINE REACTIONS. 



The use of iodine not only served to bring out certain histological peculiarities, but 

 also \'aluable data in the differentiation of different Idnds of starch. The typical or ordi- 

 narily observed reaction of starch with iodine is an indigo-blue, but if an excess of iodine 

 be avoided the reaction of the grains will be found to vary usually from a blue to a reddish- 

 violet, including within these extremes all shades of violet from a purple to a reddish-violet 

 according to the kind of starch. In fact, with few exceptions, starches are colored in the 

 presence of nunute quantities of iodine some shade of violet, varying with the kind of 

 starch. Certain starch-grains yield with any quantity of iodine a red reaction. In studying 

 this reaction we employed 0.125 and 0.25 per cent Lugol's solution. The starch was placed 

 on a sUde, and one or two or more di'ops of the iodine solution added, the whole covered with 

 a cover-slip. 



ACTIONS WITH ANILINE DYES. 



A number of these agents have been found by various investigators to be of value 

 in the differentiation of starches from different sources, of different grains of the same 

 kind of starch, and of different parts of individual grains. Quite a number of aniline dyes 

 have been used, and some experimenters have employed double or triple stains. The 

 task of selecting from these, not to consider the very large number of agents of this kind 

 that are available for such work, would of itself have been a rather large undertaking. 

 There is also no doubt that the use of double or triple stains would bring out, at times at 

 least, many points of much liistological importance, but this would have involved the 

 carrying out of the histological examinations in such detail as to be prohibitive in a research 

 of this character. Safranin and gentian violet were selected, not because they are prob- 

 ably the best of these stains for differential purposes, but because they have been found 

 very useful in starch examinations and as they yield single color reactions. 



Aniline colors in solution, especially when in weak solution ami exposed to light, are 

 notably unstable, and in order to secure strictly comparable results a quantity of a rela- 

 tively strong standard solution was prepared and kept in the dark, tightly corked. Each 

 day a certain amount of a diluted solution, containing 0.05 per cent of aniline, was pre- 

 pared from the standard solution, 5 c.c. of this was placed in a test-tube containing a small 

 amount of starch, the preparation agitated, and a drop or two of mixture withdrawn from 

 time to time, placed on a slide and covered, and the grains examined, fu-st as to the time 

 of the beginning of staining, and second as to the intensity and uniformity of coloration. 

 In the color determinations the microscope is used with low power and open diaphragm. 



REACTIONS IN POLARIZED LIGHT, WITHOUT AND WITH SELENITE. 

 Starches have been found to exhibit not only marked differences in the degrees with 

 which they rotate i)olarized light, but also differences in the characteristics of the "inter- 

 ference figure," or "cross," as it is generally termed. Moreover, this method can be used 



